Na+/H+ antiport is essential for Yersinia pestis virulence Yusuke Minato1, Amit Ghosh1†, Wyatt J ...

41 Pages·2013·0.99 MB·English

IAI Accepts, published online ahead of print on 17 June 2013
Infect. Immun. doi:10.1128/IAI.00071-13
Copyright © 2013, American Society for Microbiology. All Rights Reserved.
1 Na+/H+ antiport is essential for Yersinia pestis virulence
2
3 Yusuke Minato1, Amit Ghosh1†, Wyatt J. Faulkner2, Erin J. Lind1, Sara Schesser Bartra3,
4 Gregory V. Plano3, Clayton O. Jarrett4, B. Joseph Hinnebusch4,
5 Judith Winogrodzki5, Pavel Dibrov5 and Claudia C. Häse1,2 D
o
w
6 n
lo
a
7 1Department of Biomedical Sciences, College of Veterinary Medicine, d
e
d
8 2Department of Microbiology, College of Science, fr
o
m
9 Oregon State University, Corvallis, OR 97331, USA h
t
t
p
10 3Department of Microbiology and Immunology, :
/
/
ia
11 University of Miami, Miller School of Medicine, Miami, Florida 33101, USA i.a
s
m
12 4Laboratory of Zoonotic Pathogens, Rocky Mountain Laboratories, .
o
r
g
13 National Institute of Allergy and Infectious Diseases, National Institutes of Health, /
o
n
14 Hamilton, MT, USA. J
a
n
15 5Department of Microbiology, ua
r
y
16 University of Manitoba, Winnipeg, Manitoba R3T 2N2, Canada. 2
4
,

17 2
0
1
9
18 †Current Address: Department of Physiology, All India Institute of Medical Sciences b
y

g
19 (AIIMS), Bhubaneswar-751019, India u
e
s
20 t
21 Running Title: NhaA/B and Y. pestis virulence
22
23 Address correspondence to Claudia Häse, [email protected]
1
24 Abstract
25 Na+/H+ antiporters are ubiquitous membrane proteins that play a central role in
26 the ion homeostasis of cells. In this study, we examined the possible role of Na+/H+
27 antiport in Yersinia pestis virulence and found that Y. pestis strains lacking the major
28 Na+/H+ antiporters, NhaA and NhaB, are completely attenuated in an in vivo model of D
o
w
29 plague. The Y. pestis derivative strain lacking the nhaA and nhaB genes showed n
lo
a
30 markedly decreased survival in blood and blood serum ex vivo. Complementation of d
e
d
31 either nhaA or nhaB in trans restored the survival of the Y. pestis nhaA/nhaB double f
r
o
m
32 deletion mutant in blood. The nhaA and nhaB double deletion mutant also showed
h
t
t
33 inhibited growth in artificial serum medium, Opti-MEM® and in LB-based rich media p:
/
/
ia
34 containing similar Na+ levels and pH values as blood. Taken together, these data i.
a
s
35 strongly suggest that intact Na+/H+ antiport is indispensable for the survival of Y. pestis m
.
o
r
g
36 in the bloodstream of infected animals and thus might be regarded as a promising non- /

o
n
37 canonical drug target for infections caused by Y. pestis and possibly other blood borne J
a
n
38 bacterial pathogens. u
a
r
y
39 2
4
,
40 2
0
1
41 9
b
y

42 g
u
e
43 st
44
45
46
2
47 Introduction
48 Yersinia pestis, a Gram-negative bacterium, is the causative agent of plague.
49 Because of its high infectivity and lethality, Y. pestis is considered a potential
50 bioterrorism weapon. Y. pestis typically is introduced into the human body by the bite of
51 an infected flea and carried to the regional lymph nodes by macrophages. Following D
o
w
52 replication in the lymph nodes, the bacteria spread throughout the body by the lymph n
lo
a
53 and blood stream, and cause bubonic plague. During the later stages of the disease, d
e
d
54 the patient usually develops septicemia, disseminated intravascular coagulation, shock f
r
o
m
55 and peripheral gangrene (reviewed in (1)).
h
t
t
56 A large number of diverse bacterial species, including pathogens, use sodium as p:
/
/
ia
57 a chemiosmotic coupling ion, in addition to or even instead of protons (2-5). Sodium i.
a
s
58 motive force (SMF) may be generated either by primary sodium pumps, such as Na+- m
.
o
r
g
59 translocating NADH:ubiquinone oxidoreductases (NQR), or by secondary sodium /

o
n
60 pumps, such as Na+/H+ antiporters that are energized by the proton motive force (PMF). J
a
n
61 Although many pathogenic organisms extensively utilize different elements of the u
a
r
y
62 sodium cycle in their physiology, the role of sodium bioenergetics in bacterial virulence 2
4
,
63 in general is not well understood. Since the exposure of pathogenic microbes to the 2
0
1
64 environment of a host presents a potentially unique, and not necessarily energetically 9
b
y

65 favorable, environment for microbial growth, the capacity to use sodium as a coupling g
u
e
66 ion can be expected to be of key importance during pathogenesis in a variety of st
67 bacterial species (discussed in detail in (5)).
68 The reported genome sequence of Y. pestis predicts the presence of the primary
69 sodium pump, NQR, in addition to four secondary sodium pumps, the Na+/H+ antiporters
3
70 NhaA, NhaB, NhaC, and NhaP which are expected to contribute to SMF generation
71 across the inner membrane of Y. pestis (Fig. 1) (4). However, recent detailed
72 biochemical characterization of NhaP-type antiporters, NhaP-1 and NhaP-2, from Vibrio
73 cholerae revealed that these antiporters mediate K+/H+ exchange rather than Na+/H+
74 exchange under physiological conditions (6, 7). By analogy, we predict that the Y. D
o
w
75 pestis NhaP protein has similar cation preference (Fig. 1). Moreover, NhaC-type n
lo
a
76 antiporters are poorly characterized and Y. pestis NhaC (Yp-NhaC) remains a d
e
d
77 hypothetical protein and it is not clear at this moment whether it is chemiosmotically f
r
o
m
78 active at all. In contrast, one could expect that the NhaA-NhaB pair is the major
h
t
t
79 secondary Na+ extruding system of Y. pestis, as was found in many bacterial species p:
/
/
ia
80 possessing the nhaA and nhaB genes in their genomes (8-10). A comparison of the i.
a
s
m
81 published genomes of various Y. pestis strains (obtained from http://www.jcvi.org/ and
.
o
r
g
82 http://img.jgi.doe.gov) revealed the presence of intact genes for NhaA, NhaC and NhaP /

o
n
83 antiporters in all of the strains. Interestingly, the nhaB gene of the Y. pestis strain KIM J
a
n
84 contains a frame-shift, whereas it is intact in all of the other Y. pestis strains; thus, all Y. u
a
r
y
85 pestis KIM strains will be marked as (nhaB) to indicate the presence of this mutation. 2
4
,
86 Furthermore, genes encoding at least thirteen putative Na+-dependent symporters with 2
0
1
87 various substrate specificities and two gene orthologs of the Na+-dependent multidrug 9
b
y

88 efflux pump, NorM, are predicted as the consumers of SMF in the Y. pestis genome (Fig. g
u
e
89 1) (4). st
90 Importantly, sodium pumps are not only essential for maintaining SMF as an
91 energy source for Na+-symporters, but they also protect bacterial cells from Na+ toxicity.
92 Irrespective of the presence of functional NQR, Na+/H+ antiporters have been shown to
4
93 play a leading role in the Na+ resistance of bacteria (9, 10). Among the examined
94 Na+/H+ antiporters, antiporters of the NhaA-type are usually extremely kinetically
95 competent, fast, and pH-dependent, displaying maximal activity at alkaline pHs and are
96 indispensable for survival in sodium-rich environments, especially alkaline ones (8-10).
97 Similar to E. coli NhaA, the activity of NhaA from Y. pestis (Yp-NhaA) is regulated by pH D
o
w
98 and selectively exchanges Na+ and Li+ for H+ (11) (J. Winogrodzki and P. Dibrov, n
lo
a
99 unpublished data). NhaB-type antiporters are much less active, “housekeeping” cation d
e
d
100 exchangers that normally play an auxiliary role. Like NhaA, they exchange sodium ions f
r
o
m
101 for protons in an electrogenic manner (3 H+ per 2 Na+) and, unlike NhaA, are expressed
h
t
t
102 constitutively (9, 10, 12). Although the activity of E. coli NhaB does not show pH p:
/
/
ia
103 sensitivity (13), Vibrio alginolyticus NhaB shows pH-dependent activity, reaching its i.
a
s
m
104 maximum at alkaline pH (14). To date, no biochemical analyses of the Yp-NhaB protein
.
o
r
g
105 have been reported. /

o
n
106 Blood borne bacterial pathogens can survive in blood and thus, they can cause J
a
n
107 disease. To survive in blood, bacterial pathogens must be able to resist the multiple u
a
r
y
108 innate defense systems present in blood. In Y. pestis, a type III secretion system 2
4
,
109 (T3SS) and secreted Yop effector proteins, encoded on a Yersinia virulence plasmid, 2
0
1
110 are known to prevent phagocytosis and other protective innate immune responses. An 9
b
y

111 outer membrane protein, Ail, is also known to protect Y. pestis cells against g
u
e
112 complement-mediated lysis (15,16). In addition to the known essential factors for blood st
113 survival, Y. pestis has to be able to tolerate the salt concentrations in blood to establish
114 an infection. Thus, sodium pumps may play a role in its pathogenesis. However, to
5
115 date there is no report investigating the roles of the various Y. pestis sodium pumps in
116 the context of its pathogenic potential.
117 In this study, we investigated the roles of the primary sodium pump, NQR, and
118 the two main secondary sodium pumps, the Na+/H+ antiporters NhaA and NhaB, on Y.
119 pestis virulence. It was found that loss of both NhaA and NhaB, but not NQR, abolishes D
o
w
120 Y. pestis virulence, presumably because these antiporters are essential for protecting Y. n
lo
121 pestis from Na+ toxicity in blood. ad
e
d
122 Materials and Methods f
r
o
m
123 Bacterial strains and growth conditions. Bacterial strains and plasmids used
h
t
t
124 in this study are listed in Table 1. All bacterial strains were kept at -80°C in 20% p:
/
/
ia
125 glycerol stocks. Y. pestis strains were grown at 30°C unless otherwise noted. i.
a
s
m
126 Antibiotics were supplemented as appropriate as follows: streptomycin, 100 μg/ml;
.
o
r
127 ampicillin, 100 μg/ml; kanamycin, 50 μg/ml. g/

o
n
128 Construction of Y. pestis strains carrying deletions in nqrA-F, nhaA and J
a
n
129 nhaB. To construct the nqrA-F and nhaA deletion mutant strains of Y. pestis KIM5- u
a
r
y
130 3001 (nhaB), PCR of a KmR cassette flanked by the FRT sites and homologous regions 2
4
,

131 of the target gene was performed essentially as previously described (17) using the 2
0
1
132 primer pairs (ΔnhaA-P1 and ΔnhaA-P2, Δnqr-P1 and Δnqr-P2) listed in Table 2. The 9
b
y

133 ΔnhaA mutant was also constructed in Y. pestis KIM5 using different sets of primers gu
e
s
134 (ΔnhaA-P1B and ΔnhaA-P2B) listed also in Table 2. Y. pestis strains containing t
135 plasmid pKD46 were grown in heart infusion broth (HIB) at 28oC to an OD620 of 0.5 and
136 then for 2 h with 0.2% L-arabinose. Electrocompetent cells were prepared as previously
137 described (18) and electroporated with 2 μl of purified PCR product. Gene
6
138 replacements were confirmed by PCR analysis. The KmR cassettes were removed
139 using the temperature-sensitive plasmid pCP20, which encodes the FLP recombinase.
140 After FLP/FRT-mediated site-specific recombination, Km-sensitive colonies containing a
141 single FRT site were identified by replicate plating and PCR analysis. Temperature-
142 sensitive plasmids pKD46 and pCP20 were cured from the nqrA-F and nhaA deletion D
o
w
143 mutants by overnight growth at 38°C. The ΔnhaA and ΔnhaA/ΔnhaB double mutant n
lo
a
144 strains of Y. pestis CO92 were constructed by a similar lambda Red recombinase d
e
d
145 mutagenesis strategy (19), in which 974 bp of nhaA were replaced with a KmR cassette f
r
o
m
146 and 1453 bp of nhaB were replaced with an AmpR cassette. Primers used for these
h
t
t
147 constructs (ΔnhaA-CO-P1 and ΔnhaA-CO-P2, and ΔnhaB-CO-P1 and ΔnhaB-CO-P2) p:
/
/
ia
148 are also listed in Table 2. i.
a
s
m
149 Gene cloning of nhaA and nhaB. For the complementation analyses, we
.
o
r
g
150 cloned the nhaA gene as follows. The DNA fragment, which contained the nhaA /

o
n
151 promoter region and ORF, was amplified by PCR using chromosomal DNA of Y. pestis J
a
n
152 KIM5-3001 (nhaB) as a template. Primers used for this construct, nhaA-pWKS-P1 and u
a
r
y
153 nhaA-pWKS-P2, included SacI and BamHI restriction sites (underlined), respectively, 2
4
,

154 and were listed in Table 2. The obtained PCR products were digested with SacI and 2
0
1
9
155 BamHI, gel-purified and then ligated into the SacI and BamHI sites of the vector
b
y

156 pWKS130. pWKS130 has a pSC101 origin of replication and is stably maintained along g
u
e
s
157 with the native plasmids. The nhaB expression plasmid was constructed as follows. t
158 The nhaB ORF was amplified by using nhaB-pBAD-P1 and nhaB-pBAD-P2 (listed in
159 Table 2), and chromosomal DNA of Y. pestis KIM5-3001 (nhaB) as a template primers
160 and cloned into pBADTOPO. Because nhaB from KIM5-3001 (nhaB) contained a frame
7
161 shift mutation, we performed a site-directed mutagenesis to restore the function of the
162 gene by using primers, nhaB-pBAD-SDM-P1 and nhaB-pBAD-SDM-P2 (listed in Table
163 2), a High-Fidelity DNA Polymerase, KOD -Plus- Ver.2 (Toyobo), and Dpn I (invitrogen).
164 The site-directed mutagenesis was confirmed by sequencing. The restored nhaB was
165 amplified by PCR by using the primers, nhaB-pET-P1 and nhaB-pET-P2 (listed in D
o
w
166 Table2). The primer nhaB-pET-P2 included EcoRI restriction site (underlined). The n
lo
a
167 obtained PCR products were digested with EcoRI, gel-purified and ligated into StuI and d
e
d
168 EcoRI of the pET6xHN-C plasmid, which is located under the IPTG inducible T7 lac f
r
o
m
169 promoter.
h
t
t
170 Mouse models of septicemic and bubonic plague. Y. pestis strains were p:
/
/
ia
171 inoculated from frozen glycerol stock to 5 mL BHI and grown overnight at 28 °C without i.
a
s
m
172 aeration. Overnight cultures were diluted 1:100 into 10 mL LB and incubated again
.
o
r
g
173 overnight (16-18h) without aeration, at 37 °C. The overnight LB cultures were about /

o
n
174 1x108 cfu/mL density as determined by Petroff-Hausser chamber count and diluted in J
a
n
175 PBS. The LB cultures of the mutant and the parent strains achieved the same density. u
a
r
y
176 Eight- to 12-week-old female RML Swiss Webster mice were used. Groups of 5 -10 2
4
,
177 mice were injected intravenously in the tail vein (septicemic plague model) with 100 μl 2
0
1
178 of PBS containing 1,000 bacteria (KIM5 strains) or intradermally in the right lower 9
b
y

179 lumbar region (bubonic plague model) with 25 μl containing 10 or 100 bacteria (CO92 g
u
e
180 strains). The number of bacteria injected was confirmed by CFU count of the inoculum st
181 preparations. Infected animals were observed three times daily and euthanized upon
182 signs of terminal plague (evidence of lethargy, hunched posture, and reluctance to
183 respond to extermal stimuli). Heart blood, spleen, or both were collected immediately
8
184 after euthanasia. Dilutions of blood and triturated spleen were plated on blood agar
185 containing 1 μg/ml Irgasan to verify plague and to determine the bacterial load in these
186 tissues. These studies were conducted at biosafety level 3 and were approved by the
187 Rocky Mountain Laboratory Animal Care and Use Committee in accordance with NIH,
188 NIAID guidelines. D
o
w
189 Isolation of membrane vesicles. For the functional expression of Yp-NhaA, the n
lo
190 Na+/H+ antiporter-deficient strain of E. coli EP432, kindly provided by Dr. E. Padan ad
e
d
191 (Hebrew University of Jerusalem, Israel), was used. EP432 is a K-12 derivative, f
r
o
m
192 [melBLid, ΔnhaA1::kan, ΔnhaB1::cat, ΔlacZY, thr1] (12). EP432/pBAD-YpNhaA
h
t
t
p
193 transformants were grown in LBK medium supplemented with 100 μg/ml ampicillin, 30 :
/
/
ia
194 μg/ml kanamycin, 34 μg/ml chloramphenicol, 25 mM LiCl, and 0.01% arabinose. Cells i.a
s
m
195 were harvested at an OD600 of 1.5 to 1.8 and immediately used for isolation of inside-out .o
r
g
196 membrane vesicles as described previously (20). Briefly, overnight cultures of EP432 /
o
n

197 transformants were grown in LBK medium containing the above antibiotics. These J
a
n
u
198 cultures were then used to inoculate the growth medium at a concentration of 1:100. a
r
y

199 After harvesting, the cells were washed three times in buffer containing 200 mM sorbitol, 2
4
,

2
200 10% (w/v) glycerol and 20 mM Tris-H2SO4, pH 7.5. After the last wash, the bacterial 0
1
9
201 pellet was resuspended in the same buffer containing 1 mM 1,4-dithiothreitol (DTT), 1 b
y

g
202 μg/ml pepstatin-A, 0.1 mM phenylmethylsulfonyl fluoride (PMSF) and approximately 5 u
e
s
203 mg/L DNase. The bacterial suspension was then passed twice through a French Press t
204 (Aminco) and the unbroken cells were pelleted. The membrane fraction was collected
205 by ultracentrifugation (Optima LE-80K), and then resuspended and stored in the same
9
206 buffer containing all the additions, but without DNase and assayed for cation/proton
207 antiport activity as previously described (20-23).
208 Measurement of transmembrane ΔpH. For ΔpH measurements, aliquots of
209 vesicles (200 μg of protein) were added to 2 ml of buffer containing 200 mM sorbitol, 25
210 mM KCl, 5 mM MgCl2, 10 % (w/v) glycerol, 15 μM acridine orange and 50 mM BTP-HCl D
o
w
211 adjusted to the indicated pH. The Na+(Li+)/H+ antiport activity was then registered using n
lo
a
212 the acridine orange fluorescence quenching/dequenching assay. Respiration- d
e
d

213 dependent generation of ΔpH was initiated by the addition of 10mM Tris-D-lactate and fr
o
m
214 the resulting quenching of acridine orange fluorescence was monitored in a Shimadzu h
t
t
p
215 RF-1501 spectrofluorophotometer (excitation at 492 nm and emission at 528 nm). :/
/
ia
216 Antiport activity was estimated based on its ability to dissipate the established ΔpH in i.a
s
m
217 response to the addition of NaCl or LiCl at the indicated concentrations. 10 mM of each .o
r
g
/
218 was used in the determination of the pH profile of activity and 0.1 mM to 20 mM was o
n

J
219 used in the determination of apparental Km. The antiport activities are expressed as a
n
u
220 percent restoration of lactate-induced fluorescence quenching. All assays were a
r
y

2
221 performed at room temperature. 4
,

2
222 Blood and blood serum survival assay. Defibrinated sheep blood and sheep 0
1
9

223 serum (Colorado Serum Company) were used for the animal blood and serum survival b
y

g
224 assays. Human blood was purchased from Biological Specialty Corporation. Human u
e
s
t
225 serum was purchased from Lonza. Bacterial strains were grown overnight in LB
226 medium at 30°C, washed by PBS, and suspended in PBS. Approximately 105 or 106
227 colony forming units (CFU) of the diluted bacteria were inoculated into 4 ml of the blood
228 or blood serum and shaken for 18 hours at 30°C. CFU were determined before and
10